This proposal can be divided into three sections. The first, is direct extension of our previous work on genomic instability in tumor progression. We had measured the rate of spontaneous mutation and of the generation of karyotypic changes in metastatic and non-metastatic cells but could not demonstrate a difference in either parameter between the cell types. However, because the parameters we chose may have been to finite (point mutations) or to broad (cytogenetic changes), we have chosen to look at homologous rearrangements (HR) as a means of determining if significant differences in HR for metastatic and non-metastatic cells can be detected. The second section of this proposal extends our interest in progression to the issue of clonal dominance in relationship to the development of progressed, i.e. metastatic cells. Recent reports have indicated (in a murine model) that as tumors become metastatic a single clone dominates not only the primary tumor, but also the metastases. We will use molecular techniques to address this issue in human tumors and determine whether clonal dominance is a feature of human neoplasia. Since we believe the mechanism by which clonal dominance occurs relates to genomic instability (please see text), these analyses establish a reasonable link between genomic instability and progression. This issue will also be analyzed in a carcinogenesis model during the conversion of papillomas to carcinomas. The third section of this proposal deals with our hypothesis that tumor progression is of two types, namely the classic form as seen in colon carcinoma (Type 1) and the accelerated form, (Type 2), which occurs in most other tumors. We have chosen unknown primary tumors as being the quintessential example of a Type 2 tumor progressor. These studies are aimed at defining Type 2 tumors using karyotypic analyses in an effort to define unique chromosomal changes in these tumors. We also plan to use the approaches contained in the first two sections to study the relationship between Type 1 and Type 2 progressors. We are, therefore, proposing a continuation of our earlier work and Its extension to newer areas using molecular genetic techniques to assess genomic instability and tumor progression.